Force on ForcePrime Rogue Inc.
A grognard exercise, not a scientific model. What that means

Modules

What is in the game. Each module can be switched off, and is measured by playing the same 3,000 wars with it on and with it off.

Every parameter carries a source line, and the version that added it. A judgement tag marks a number that is the modeller's estimate, not a measurement: 173 of 224. Three numbers in a row are low, most likely and high; each run draws its own value between them.

The base game

0.1CMBConventional fight 

Conventional fight

The opening campaign, resolved daily. Combat power is people times effectiveness times air superiority; loss rates follow the power ratio. Backtested against the 2003 invasion of Iraq.

Switch always onVersions 0.1Parameters 7
Parameters and their sources (7)
ParameterValue or low – likely – highUnitSourceAdded in
kinetic.attrition_base_rate0.006 – 0.01 – 0.016fraction of committed force becoming casualties per day at power ratio 1:1Dupuy, Attrition (1990): division-level daily casualty rates ~1-2%/day in intense combat; calibrated vs Iraq 20030.1
kinetic.ratio_exponent1.5 – 2 – 2.5elasticity of attacker casualty rate to (defender/attacker) power ratio; defender uses half this elasticityBiddle, Military Power (2004): modern-system force employment produces highly lopsided exchange ratios; calibrated vs Iraq 2003 (calibration/iraq_2003_invasion.yaml)0.1
kinetic.defender_posture_mult1.2 – 1.45 – 1.8multiplier on defender combat powerDupuy QJM posture factors (hasty 1.3, prepared 1.5, fortified 1.6+)0.1
kinetic.kia_fraction0.17 – 0.22 – 0.28killed / total casualtiesUS historical KIA:WIA ratios (Iraq/Afghanistan ~1:7 to 1:3); modeller rangejudgement0.1
kinetic.defender_breakpoint0.25 – 0.35 – 0.5cumulative casualty fraction at which organized conventional defence collapsesDupuy / Bracken breakpoint literature: units typically break at 20-40% losses; national armies higher0.1
kinetic.pursuit_rate_factor0.005 – 0.012 – 0.025attacker casualty rate after defender breaks, as a fraction of attrition_base_rateCalibrated vs Iraq 2003 (advance to Baghdad after Iraqi conventional collapse)0.1
kinetic.air_superiority_mult1.5 – 2 – 3combat-power multiplier for side holding air superiorityDupuy (QJM air factors); 1991/2003 Iraq experience; modeller rangejudgement0.1
0.1OCCOccupation and resistance 

Occupation and resistance

Resolved monthly. A resistance index, a troop requirement that follows Quinlivan's force ratios, control as troops deployed over troops needed, occupier deaths, and a monthly chance the insurgency ends.

Switch always onVersions 0.1Parameters 10
Parameters and their sources (10)
ParameterValue or low – likely – highUnitSourceAdded in
occupation.density_high_resistance20troops per 1000 populationQuinlivan, 'Force Requirements in Stability Operations', Parameters 19950.1
occupation.density_low_resistance2.5troops per 1000 populationQuinlivan 1995 (permissive environments ~1-4 per 1000)0.1
occupation.hostile_kia_rate_ref9e-05 – 0.00013 – 0.00018occupier hostile deaths per troop-month at reference resistance R_ref, before shortfall penaltyCalibrated so the Iraq 2003-11 backtest (calibration/iraq_2003_occupation) reproduces ~3,500 US hostile deaths over ~104 months at ~140k troops; DoD DCAS0.1
occupation.resistance_relaxation0.05 – 0.08 – 0.12monthly rate at which resistance moves toward its equilibriumModeller estimate; to calibrate on monthly Iraq SIGACTSjudgement0.1
occupation.shortfall_resistance_gain0.25 – 0.4 – 0.55equilibrium resistance added per unit of control shortfallModeller estimate (Quinlivan's mechanism: too few troops -> insecurity -> more insurgency)judgement0.1
occupation.resistance_ref0.6resistance index of the reference case (Iraq 2004-07)Modeller anchorjudgement0.1
occupation.insurgency_median_months72 – 120 – 168monthsConnable & Libicki, How Insurgencies End (RAND 2010): median ~10 years0.1
occupation.external_support_resistance_boost0.1 – 0.2 – 0.3additive boost to resistance index per support level above 'diplomatic'Connable & Libicki 2010; Byman et al. (RAND 2001): external support strongly predicts insurgent durability0.1
occupation.local_force_substitution0 – 0.15 – 0.35fraction of density requirement met by collaborating local police/administrationModeller estimate; Iraq/Afghanistan local-force build-up timelinesjudgement0.1
occupation.civil_resistance_cost_mult0.5 – 1 – 1.5scales GDP loss from non-cooperation (strikes, administrative refusal)Chenoweth & Stephan, Why Civil Resistance Works (2011) — qualitative; modeller scalingjudgement0.1
0.1POLPolitics in both capitals 

Politics in both capitals

US public support falls with deaths and economic loss and sets a monthly chance of withdrawal, higher in election months. Canada's government has a monthly chance of capitulating. That second hazard is the least calibrated part of the model.

Switch always onVersions 0.1Parameters 7
Parameters and their sources (7)
ParameterValue or low – likely – highUnitSourceAdded in
political.mueller_slope0.06 – 0.1 – 0.15support drop (fraction) per 10x cumulative own KIAMueller, War, Presidents and Public Opinion (1973); Gartner & Segura 19980.1
political.casualty_reference25 – 50 – 100KIA scale in log10(1 + KIA/ref); damps the over-steep early drop of a pure log curveFitted by eye to Iraq 2003-08 Gallup support vs DoD cumulative deaths (~-0.25 by 1,000, ~-0.35 by 4,000 incl. non-casualty factors); to fit formally0.1
political.econ_slope0.01 – 0.025 – 0.04support drop per 1% GDP lossModeller estimate (economic-voting elasticities); to calibratejudgement0.1
political.withdrawal_threshold0.15 – 0.22 – 0.3public support below which withdrawal pressure becomes acuteModeller estimate; Vietnam/Iraq withdrawal support levelsjudgement0.1
political.withdrawal_max_hazard0.01 – 0.025 – 0.05monthly probability of termination between elections when support is far below thresholdExecutives sustain unpopular wars for years (Iraq 2006-08: support ~30%, no withdrawal until the 2008 SOFA); modeller estimatejudgement0.1
political.election_hazard_mult3 – 5 – 8multiplier on withdrawal hazard in a national election monthModeller estimate; elections are the main institutional exit point (1968, 2008)judgement0.1
political.capitulation_base_hazard0.002 – 0.008 – 0.02monthly probability defender government accepts terms, scaled by losses / resolveUNCALIBRATED modeller estimate. Target for calibration: Hufbauer-Schott-Elliott (~1/3 sanctions success for major goals) vs Pape (<5%); no modern case of a democracy ceding sovereignty to coercionjudgement0.1
0.1ECNEconomy 

Economy

Lost trade, energy shocks, sanctions and war damage as GDP loss on both sides, with partial recovery as the occupier gains control.

Switch always onVersions 0.1Parameters 4
Parameters and their sources (4)
ParameterValue or low – likely – highUnitSourceAdded in
economic.trade_va_share0.45 – 0.6 – 0.75domestic value-added share of trade flows lost when trade is disruptedOECD TiVA order of magnitude; to cite exact Canada-US figures0.1
economic.supply_chain_amplifier1.2 – 1.6 – 2.2multiplier on direct trade loss for integrated supply chains (autos, energy)Modeller estimate; ~50% of bilateral goods trade is intra-firm (analyzify summary of USTR/StatCan)judgement0.1
economic.oil_shock_gdp_per_mbd0.1 – 0.2 – 0.35% US GDP loss per 1 Mb/d crude supply lost (short run)Modeller estimate from oil-shock elasticity literature (Hamilton); to calibratejudgement0.1
economic.allied_sanction_gdp_hit0.2 – 0.6 – 1.5% US GDP loss when allies impose sanctionsModeller estimatejudgement0.1

Forces

0.1.1MOBCitizen mobilization+4.5+7.5

Citizen mobilization

Canada's Defence Mobilization Plan (a Primary Reserve of 100,000 and a Supplementary Reserve of 300,000) and full US mobilization with a draft. Mobilized Canadians join the opening fight, seed the resistance, and make the country harder to talk into surrender.

Switch canada_mobilization.modeVersions 0.1.1, 0.1.3.6, 0.1.3.7Parameters 16

What it is worth now

ScenarioOutcome countedPoints, module on against off95% interval
Full invasionAmericans go home+4.5+3.8 to +5.3
Alberta onlyAmericans go home+7.5+6.6 to +8.4
No invasionUS abandons the campaign+0.8+0.5 to +1.1

Everything else on. 3,000 paired runs, v0.1.3.8. Zero in the no-invasion scenario usually means the module has no path there.

Parameters and their sources (16)
ParameterValue or low – likely – highUnitSourceAdded in
mobilization.primary_fill0.3 – 0.5 – 0.8fraction of the 100,000 Primary Reserve target actually in uniform when war starts (today 23,561 = 0.24)Modeller estimate; DMP gives no timeline and CAF converts only ~1 in 13 applicants (NATO Association of Canada 2026)judgement0.1.1
mobilization.supplementary_fill0.15 – 0.4 – 0.75fraction of the 300,000 Supplementary Reserve target enrolled when war starts or reachable within weeksModeller estimate; DMP is in design (Globe and Mail); wartime volunteer surge (Ukraine 2022) supports the upper rangejudgement0.1.1
mobilization.arms_fraction0.2 – 0.4 – 0.7fraction of mobilized supplementary personnel Canada can armModeller estimate; DND documents cite equipment shortages and supply-chain limits (Vincent Rigby, CBC, Dec 2025); 30,000 new rifles ordered from Colt Canada (2026)judgement0.1.1
mobilization.primary_cev0.45 – 0.6 – 0.75combat effectiveness of newly expanded Primary Reserve relative to CAF regulars (=1 of defender CEV)Modeller estimatejudgement0.1.1
mobilization.supplementary_cev0.1 – 0.2 – 0.3combat effectiveness of supplementary reservists (one week of training) relative to CAF regularsDMP: about one week of training in shooting, moving, communicating, driving and drones (CTV; NPR). Modeller estimatejudgement0.1.1
mobilization.primary_arrival_day3 – 7 – 14days until mobilized Primary Reserve units are in the fightModeller estimatejudgement0.1.1
mobilization.supplementary_arrival_day10 – 21 – 45days until armed supplementary reservists are organized and in the fightModeller estimatejudgement0.1.1
mobilization.conventional_commit_share0.3 – 0.5 – 0.7share of armed mobilized personnel committed to the conventional fight (rest held for territorial defence / resistance)Modeller estimatejudgement0.1.1
mobilization.cadre_resistance_coef0.02 – 0.04 – 0.06resistance index added per trained, armed survivor per 1,000 populationModeller estimate; organized pre-war cadres underpin durable resistance (Norway, Baltics 1944-53; Ukraine TDF 2022)judgement0.1.1
mobilization.cadre_resistance_cap0.25maximum resistance index added by cadresModeller capjudgement0.1.1
mobilization.resolve_capitulation_damping0.1 – 0.25 – 0.4proportional cut in defender capitulation hazard when organized mobilization is in place (scaled by fill)Modeller estimate (whole-of-society defence raises the cost of surrender; Finland model cited in the DMP)judgement0.1.1
mobilization.us_guard_mobilized_full0.3 – 0.45 – 0.6fraction of Guard/Reserve mobilized under full mobilization (vs 5-25% standard)Modeller estimate; ~1/3 of Guard/Reserve mobilized at peak of Iraq/Afghanistanjudgement0.1.1
mobilization.us_draft_month4 – 6 – 10month the draft is enacted (only if occupation control < 0.8)Modeller estimate; requires Congress to restore induction authority under the Military Selective Service Actjudgement0.1.1
mobilization.us_draft_inductees150,000 – 300,000 – 500,000inductees available for occupation duty once trainedModeller estimate (Vietnam peak ~300k inductions/yr)judgement0.1.1
mobilization.us_draft_training_months4 – 6 – 9months from induction to deployabilityBasic + AIT ~ 4-6 months; modeller rangejudgement0.1.1
mobilization.us_draft_support_penalty0.05 – 0.1 – 0.15one-time drop in US public support when the draft is enactedModeller estimate; Vietnam-era draft oppositionjudgement0.1.1
Data: data/actors/canada.yaml (16 entries)
EntryValueUnitAs ofSourceConfidenceAdded in
population41,798,407persons2026-07-01Statistics Canada, Table 17-10-0009-01, preliminary estimate for 1 July 2026 (released 23 September 2026); was 41.6e6 (2025) before v0.1.3.1high0.1
gdp_usd2.2 trillionUSD (nominal)2025CalChamber Canada trade portal (citing StatCan)medium0.1
defence_spending_cad63 billionCAD/yrFY2025-26NATO defence expenditure report, 26 Mar 2026 (reported by National Post / Global News); ~2.0% GDP under broadened NATO definitionhigh0.1
regular_force67,827personnel2026-04DND/CAF backgrounder 'CAF reach highest recruitment in 30 years', 20 Apr 2026high0.1
regular_force_target71,500personnel2029 targetDND/CAF backgrounder, 20 Apr 2026high0.1
reserve_force30,000personnel (authorized)2026DND/CAF backgrounder, 20 Apr 2026 (authorized strength; actual effective strength lower)medium0.1
land_combat_share0.35fraction of regular force available as deployable land combat/combat-support2026Modeller estimate — Army is ~1/3 of CAF; to replace with DND Departmental Results Report breakdownestimate0.1
mobilization_plan_regular_augment85,000personnel (proposed)2025-12NATO Association of Canada, reporting on DND/CAF mobilization planning (proposal, not enacted)low0.1
trade.exports_to_us_usd453.6 billionUSD/yr (goods + services, = US imports from Canada)2025BEA Table 1.5 via USAFactshigh0.1
trade.imports_from_us_usd426.3 billionUSD/yr (goods + services)2025BEA Table 1.5 via USAFactshigh0.1
trade.crude_exports_to_us_bpd3.9 millionbarrels/day2025EIA Today in Energy #67904, 29 Jul 2026high0.1
primary_reserve_actual23,561personnel2025-05DMP directive figure (reported by Globe and Mail, The Deep Dive, True North)high0.1.1
primary_reserve_dmp_target100,000personnelDMP target (no date set)DMP directive (Globe and Mail; CTV/BNN Bloomberg 20 Nov 2025)high0.1.1
supplementary_reserve_actual4,384personnel (mostly retired CAF willing to return)2025-05DMP directive (Western Standard; The Deep Dive); still 'the current 4,384' in Ottawa Citizen, 16 Nov 2026
DND's Reserve Force components page says 'about 6,700' (undated). No source found for 4,135 (checked 17 Nov 2026). The difference has no effect on results (v0.1.3.6).
high0.1.1
supplementary_reserve_dmp_target300,000personnel (volunteers, initially public servants; relaxed age/fitness standards; ~1 week training; no uniforms)DMP target (emergency)DMP directive (CTV/BNN Bloomberg 20 Nov 2025; NPR 15 Dec 2025; True North)high0.1.1
caf_applicants_fy2025_26192,000applicants (1 in 13 completed the process, ~15,000 recruits)FY2025-26NATO Association of Canada, 'Modernizing Canada's Reserve Force' (2026)medium0.1.1
Data: data/actors/united_states.yaml (8 entries)
EntryValueUnitAs ofSourceConfidenceAdded in
population340 millionpersons2025Modeller estimate (Census Bureau order of magnitude) — replace with Census Vintage 2025estimate0.1
gdp_usd30 trillionUSD (nominal)2025Modeller estimate (BEA order of magnitude) — replace with BEA annual 2025estimate0.1
active_end_strength1.3 millionpersonnelFY2026 NDAAMilitary Times, 20 Jan 2026; CRS IN12651 (FY2026 NDAA active end strength)high0.1
army_active454,000personnelFY2026Military Times, 20 Jan 2026high0.1
marines_active172,300personnelFY2026Military Times, 20 Jan 2026high0.1
reserve_and_guard772,000personnelFY2026Military Times, 20 Jan 2026; CRS IN12652high0.1
sustainable_deploy_fraction0.33fraction of active ground force sustainable in theatre without mobilization (1:2 rotation)doctrineModeller estimate based on Army force-generation dwell-ratio norms; to citeestimate0.1
trade.crude_imports_from_canada_share0.6fraction of US crude imports2025EIA company-level imports 2025 (Canada ~3.8-4.0 Mb/d of ~6 Mb/d total)medium0.1

Environment

0.1.2WXWeather−0.2+0.4

Weather

Daily weather in five regions built from 1991 to 2020 climate normals, with cold snaps that persist, weighted by where the fighting is. Cold costs an occupier strength and casualties according to its cold-weather proficiency.

Switch weather.enabled, weather.modelVersions 0.1.2, 0.1.2.6Parameters 27

What it is worth now

ScenarioOutcome countedPoints, module on against off95% interval
Full invasionAmericans go home−0.2 (within noise)−0.6 to +0.1
Alberta onlyAmericans go home+0.4+0.1 to +0.6
No invasionUS abandons the campaign+0.1 (within noise)−0.0 to +0.2

Everything else on. 3,000 paired runs, v0.1.3.8. Zero in the no-invasion scenario usually means the module has no path there.

Parameters and their sources (27)
ParameterValue or low – likely – highUnitSourceAdded in
weather.temp_anomaly_sd1.5 – 2.5 – 3.5°C, standard deviation of the per-run (whole-war) monthly temperature anomalyModeller estimate; interannual SD of Canadian winter monthly means is roughly 2-4 °C (ECCC station records)judgement0.1.2
weather.snow_anomaly_sigma0.35lognormal sigma on monthly snowfall for the runModeller estimatejudgement0.1.2
weather.mobility_penalty0.3 – 0.6 – 1proportional lengthening of the conventional campaign at severity W = 1Modeller estimate; winter campaigns slow sharply (Winter War 1939-40; Russia-Ukraine winter tempo 2022-24)judgement0.1.2
weather.air_penalty0.15 – 0.3 – 0.45share of the air-superiority advantage lost at W = 1 (cloud, icing, short days)Modeller estimate; Montréal is overcast or mostly cloudy ~64% of the time in January (WeatherSpark)judgement0.1.2
weather.defender_bonus0 – 0.1 – 0.2added defender posture multiplier at W = 1 (home-adapted troops in prepared positions)Modeller estimate; Finland 1939-40judgement0.1.2
weather.cold_injury_rate0.0002 – 0.0005 – 0.001attacker non-battle cold casualties per soldier per day at cold = 1 (-20 °C mean)Modeller estimate; cold-weather injury burdens in Korea 1950-51 and Arctic exercisesjudgement0.1.2
weather.insurgent_winter_lull0.1 – 0.25 – 0.4proportional drop in occupier hostile deaths at W = 1Modeller estimate; Afghanistan 'fighting season' seasonalityjudgement0.1.2
weather.heating_mult0.05 – 0.1 – 0.2proportional increase in economic pain from disruption at cold = 1 (heating season)Modeller estimate; to be replaced by the gas/heating set (0.1.12)judgement0.1.2
weather_complex.daily_persistence0.7 – 0.8 – 0.88lag-1 autocorrelation of daily temperature anomalies (cold snaps and warm spells persist)Modeller estimate; typical mid-latitude daily anomaly persistencejudgement0.1.2.6
weather_complex.daily_sd_winter4.5 – 6 – 7.5°C, SD of daily temperature anomaly Nov-Mar (continental Canada)Modeller estimatejudgement0.1.2.6
weather_complex.daily_sd_summer2 – 3 – 4°C, SD of daily temperature anomaly May-SepModeller estimatejudgement0.1.2.6
weather_complex.enso_regimela_nina 0.25, neutral 0.5, el_nino 0.25ENSO state for the war's wintersENSO events recur every 2-7 years (NOAA via Global News 2024); shares are a modeller estimatejudgement0.1.2.6
weather_complex.enso_winter_offset_west1 – 1.5 – 2.5°C winter (Nov-Mar) offset in prairies/oil sands/north/pacific; negative under La Nina, positive under El NinoEnvironment Canada: La Nina brings colder-than-normal Prairie winters; El Nino warmer and drier West (Western Producer 2022, 2024); AccuWeather 2021-22: 1.5-2 °C below average southern Prairies0.1.2.6
weather_complex.enso_winter_offset_east0.5°C winter offset in the corridor (weaker signal; La Nina wetter in Ontario/Quebec)Environment Canada via Western Producer (2022)0.1.2.6
weather_complex.warming_offset0.3 – 0.7 – 1.2°C by which 2026 conditions exceed the 1991-2020 normals (annual); x1.5 in winter, x1.3 in the northCanada warmed 2.0 °C over 1948-2023, mostly since 1970, north 2.6 °C, more in winter (CCCR 2026 ch.2); offset relative to the 2005-centred normals is a modeller estimatejudgement0.1.2.6
weather_complex.cold_proficiency_attacker0.25 – 0.4 – 0.55share of cold-weather penalties the US force avoids (training, kit, units); 0.5 = the simple model's implicit level11th Airborne is the US Army's only Arctic-capable division (USU 2026; Bloomberg 2026); rest of force largely untrained for extreme cold; modeller estimatejudgement0.1.2.6
weather_complex.cold_proficiency_defender0.55 – 0.7 – 0.85share of cold-weather penalties Canadian forces avoid (home terrain, winter warfare standard, Rangers)Modeller estimatejudgement0.1.2.6
weather_complex.air_denied_penalty0.4 – 0.6 – 0.8share of the air-superiority advantage lost on a fully weather-denied day (blizzard, freezing rain, heavy smoke)Modeller estimatejudgement0.1.2.6
weather_complex.darkness_penalty0.05 – 0.1 – 0.2share of air advantage lost at 0 h of daylight relative to 12 h (night-capable forces limit this)Modeller estimatejudgement0.1.2.6
weather_complex.blizzard_cover0 – 0.15 – 0.3added insurgent effectiveness on blizzard days (cover, degraded surveillance)Modeller estimatejudgement0.1.2.6
weather_complex.cold_snap_lull0.2 – 0.4 – 0.6drop in insurgent tempo on extreme-cold daysModeller estimatejudgement0.1.2.6
weather_complex.occupation_cold_strength_loss0.002 – 0.005 – 0.01share of occupier strength lost per extreme-cold day to cold injuries and kit failure, scaled by (1 - proficiency)Modeller estimate; Army doctrine notes cold casualties can rival battle casualties (Task & Purpose, 2025)judgement0.1.2.6
weather_complex.ice_storm_prob0.01 – 0.03 – 0.06chance per winter of a 1998-scale multi-day ice storm in the corridor1998: 80+ hours of freezing rain vs 45-65 h normal per year; 4 other major events 1958-1998 (Theor. Appl. Climatol. 2000)0.1.2.6
weather_complex.ice_storm_can_gdp0.3 – 0.6 – 1.2% Canadian GDP lost in an ice-storm monthModeller estimate; 1998 storm left 1.4M Quebec customers without power, last restored 6 Febjudgement0.1.2.6
weather_complex.fire_season_sigma0.6lognormal sigma of each fire season's severity multiplier on smoke days and fire risk (2023 ~17M ha, 2025 ~8.9M ha vs ~4.2M ha 10-yr avg)CIFFC via The Weather Network and This Magazine (2025)0.1.2.6
weather_complex.oil_sands_fire_prob0.02 – 0.05 – 0.1chance per fire season (at severity 1) that wildfire shuts in oil sands output for about a month2016 Fort McMurray fire: ~0.8 Mb/d lost on average in May, peak >1.1 Mb/d (EIA, June 2016)0.1.2.6
weather_complex.oil_sands_fire_mbd0.6 – 0.9 – 1.2Mb/d shut in during an oil sands fire monthEIA (June 2016); NEB (May 2016)0.1.2.6
Data: data/climate/canada_stations.yaml (8 entries)
EntryValueUnitAs ofSourceConfidenceAdded in
stations.montrealmonthly normalsECCC Canadian Climate Normals 1991-2020, via Template:Montréal–Trudeau weatherbox0.1.2
stations.torontomonthly normalsECCC Canadian Climate Normals 1991-2020, via Template:Pearson Airport weatherbox0.1.2
stations.winnipegmonthly normalsECCC Canadian Climate Normals 1991-2020, via Template:Winnipeg weatherbox0.1.2
stations.calgarymonthly normalsECCC Canadian Climate Normals 1991-2020, via Template:Calgary weatherbox0.1.2
stations.vancouvermonthly normalsECCC Canadian Climate Normals 1991-2020, via Wikipedia 'Climate of Vancouver'0.1.2
stations.iqaluitmonthly normalsECCC Canadian Climate Normals 1991-2020, via Template:Iqaluit weatherbox0.1.2
stations.fort_mcmurraymonthly normalsECCC Canadian Climate Normals 1991-2020, via Template:Fort McMurray weatherbox (fetched 11 Nov 2026)0.1.2.6
stations.yellowknifemonthly normalsECCC Canadian Climate Normals 1991-2020, via Template:Yellowknife weatherbox (fetched 11 Nov 2026)0.1.2.6
0.1.2.2SEASeasonal insurgency0+0.2

Seasonal insurgency

The resistance has a fighting season. Tempo, pressure on occupier troops and mud seasons vary through the year.

Switch seasonal_insurgency.enabledVersions 0.1.2.2Parameters 2

What it is worth now

ScenarioOutcome countedPoints, module on against off95% interval
Full invasionAmericans go home+0.0 (within noise)−0.2 to +0.2
Alberta onlyAmericans go home+0.2+0.0 to +0.3
No invasionUS abandons the campaign+0.0 (within noise)+0.0 to +0.0

Everything else on. 3,000 paired runs, v0.1.3.8. Zero in the no-invasion scenario usually means the module has no path there.

Parameters and their sources (2)
ParameterValue or low – likely – highUnitSourceAdded in
seasonal.profile_amplitude0.5 – 1 – 1.5scales the monthly insurgent-tempo profile (1 = base profile, peak-to-trough ~1.9:1)Modeller estimate; Afghanistan fighting season (spring-fall peaks, winter lulls in ISAF SIGACTS); Baltic 'Forest Brothers' 1944-53 summer activity and winter bunkersjudgement0.1.2.2
seasonal.mud_season_penalty0.1 – 0.25 – 0.4proportional lengthening of movement in spring break-up (April); 60% of that at fall freeze-up (November)Modeller estimate; rasputitsa effects in Ukraine 2022-23; closure of Canadian winter roads at break-upjudgement0.1.2.2
0.1.2.6.1WLDBears and wildlife0+0.9

Bears and wildlife

Collisions with large animals, bear encounters and summer insects as a small, constant drain on an occupier, and polar bears as a story that can move opinion.

Switch wildlife.enabledVersions 0.1.2.6.1Parameters 10

What it is worth now

ScenarioOutcome countedPoints, module on against off95% interval
Full invasionAmericans go home−0.0 (within noise)−0.1 to +0.0
Alberta onlyAmericans go home+0.9+0.5 to +1.3
No invasionUS abandons the campaign+0.0 (within noise)+0.0 to +0.0

Everything else on. 3,000 paired runs, v0.1.3.8. Zero in the no-invasion scenario usually means the module has no path there.

Parameters and their sources (10)
ParameterValue or low – likely – highUnitSourceAdded in
wildlife.vehicle_km_per_troop_day15 – 25 – 40vehicle-km driven per occupier per day (patrols, convoys, resupply)Modeller estimatejudgement0.1.2.6.1
wildlife.collision_rate_per_mkm0.1 – 0.2 – 0.35large-animal collisions per million vehicle-km on rural/night military routes (civilian average ~0.1-0.2, raised for night and rural exposure)Modeller estimate from 35-70k collisions a year on ~330 billion vehicle-km (Transport Canada)judgement0.1.2.6.1
wildlife.collision_fatality0.0005 – 0.001 – 0.002deaths per large-animal collision (civilian ~27 a year over tens of thousands of collisions; moose far deadlier)TIRF 2000-2020 via eDriving; modeller estimatejudgement0.1.2.6.1
wildlife.collision_injury0.03 – 0.07 – 0.1injuries per collision~7% of wildlife crashes involve injury or death (Transport Canada via animal-world.com)0.1.2.6.1
wildlife.bear_encounter_rate0.5 – 1.5 – 4dangerous bear encounters per 10,000 troops in bear country per active month (Apr-Nov; polar bears year-round)Modeller estimatejudgement0.1.2.6.1
wildlife.bear_casualty_share0.02 – 0.05 – 0.1share of dangerous encounters that injure someone (armed troops usually shoot first); ~15% of attacks fatal (Wilder et al. 2017)Modeller estimate; Wilder et al. (2017)judgement0.1.2.6.1
wildlife.polar_bear_guard_overhead0.03 – 0.06 – 0.1share of occupier troops in polar-bear range tied up on bear watch and camp securityModeller estimate; Arctic operations require armed bear monitors (Canadian Rangers' role)judgement0.1.2.6.1
wildlife.insect_penalty0.02 – 0.05 – 0.1loss of occupier effectiveness in boreal regions in Jun-Aug (blackflies, mosquitoes); locals are acclimatizedModeller estimatejudgement0.1.2.6.1
wildlife.iconic_incident_prob0.02 – 0.05 – 0.1monthly chance, with US troops in polar-bear range, that troops kill a polar bear in a widely shared incidentModeller estimate; defence-of-life kills occur in Churchill and Nunavut most yearsjudgement0.1.2.6.1
wildlife.iconic_incident_effect-0.005 – 0.003 – 0.01lasting drop in US support after such an incident (negative = no harm or slight rally)Modeller estimate; belongs in the narrative module (0.1.21b)judgement0.1.2.6.1
Data: data/commodities/wildlife.yaml (5 entries)
EntryValueUnitAs ofSourceConfidenceAdded in
wildlife_vehicle_collisions_per_hour4 to 8large-animal vehicle collisions per hour in Canada (~35,000-70,000 a year); ~80% deer2011Transport Canada via State Farm / Canadian Underwriter (2011); BC Wildlife Collision Prevention Programmedium0.1.2.6.1
wildlife_vehicle_deaths570people killed in wildlife-vehicle collisions in Canada 2000-2020 (~27 a year); over half moose, ~32% deer; 42% Jun-Aug, 33% Sep-Nov2020Traffic Injury Research Foundation via eDriving (2024)high0.1.2.6.1
wildlife_collisions_dark_share0.8share of commercial-fleet wildlife collisions in darkness (8 PM-9 AM); Trans-Canada Highway the top corridor2026Samsara fleet data, April 2026medium0.1.2.6.1
polar_bear_attacks73 attacks 1870-2014, 20 deaths, 63 injuries across all five range states; most predatory; most attacks in 2010-2014 (low sea ice)text2017Wilder et al. (2017), Wildlife Society Bulletin 41(3)high0.1.2.6.1
bear_attacks_north_americaabout 11 attacks a year; grizzly ~1.6 and black bear ~1.25 deaths a year in North Americatext2017Secondary compilations (animal-world.com, petpedia) of published studies; low confidencelow0.1.2.6.1

Economy

0.1.2.3POTPotash0+2.5

Potash

Canada supplies most US potash. Losing it is a slow farm shock, eased over time by other suppliers. Canada can embargo it; the US can take the basin.

Switch potash.enabledVersions 0.1.2.3Parameters 8

What it is worth now

ScenarioOutcome countedPoints, module on against off95% interval
Full invasionAmericans go home+0.0 (within noise)+0.0 to +0.0
Alberta onlyAmericans go home+2.5+1.9 to +3.1
No invasionUS abandons the campaign+11.5+10.3 to +12.6

Everything else on. 3,000 paired runs, v0.1.3.8. Zero in the no-invasion scenario usually means the module has no path there.

Parameters and their sources (8)
ParameterValue or low – likely – highUnitSourceAdded in
potash.substitution_max0.25 – 0.45 – 0.65share of lost Canadian potash the US can replace from other suppliers once markets adjustModeller estimate. Russia/Belarus are the only large alternatives (sanctions; a US-Belarus purchase deal was floated in Sept 2026, Railway Age); Israel, Jordan, Germany, Laos are smaller and committedjudgement0.1.2.3
potash.substitution_months6 – 12 – 18months to reach maximum substitutionModeller estimate; contract and shipping cyclesjudgement0.1.2.3
potash.occupied_mine_output0.3 – 0.6 – 0.8share of Saskatchewan output that flows south under full US control (workforce, rail and sabotage limits)Modeller estimatejudgement0.1.2.3
potash.political_weight0.03 – 0.06 – 0.1drop in US public support when US farmers face a full potash shortfall (level, not cumulative)Modeller estimate; farm-state political salience of 2022 fertilizer price spike (USDA AMS comment record)judgement0.1.2.3
potash.gdp_weight0.05 – 0.1 – 0.2% US GDP lost at a full potash shortfall (farm incomes, food prices, lagged yields)Modeller estimate; fertilizer ~15% of farm cash costs (USDA), potash prices +178% in 2021-22 (World Bank)judgement0.1.2.3
potash.application_season_mult1.4multiplier on political effect in spring (Mar-May) and fall (Sep-Nov) application months; 0.6 otherwiseModeller estimatejudgement0.1.2.3
potash.offshore_diversion0.1 – 0.25 – 0.4share of US-bound Canadian potash that can be re-sold offshore (Neptune/Vancouver capacity; Portland route lost in war)Modeller estimate; Canpotex terminal mix (Railway Age 2026)judgement0.1.2.3
potash.cad_usd0.72USD per CADApproximate 2025-26 rate0.1.2.3
Data: data/commodities/potash.yaml (10 entries)
EntryValueUnitAs ofSourceConfidenceAdded in
us_net_import_reliance0.92share of US apparent consumption imported2025USGS Mineral Commodity Summaries 2026, Potashhigh0.1.2.3
us_imports_from_canada_share0.79share of US potash imports from Canada (2021-24 basis)2025USGS Mineral Commodity Summaries 2026, Potash (Canada 79%, Russia 12%, Israel 3%, other 6%)high0.1.2.3
us_consumption_k2o_t5.9 milliontonnes K2O, apparent consumption2025USGS MCS 2026, Potash (as reported by theworlddata.com summary)medium0.1.2.3
canada_production_k2o_t15 milliontonnes K2O (~31% of world mine production)2025USGS MCS 2026 via Fertilizer Daily, 29 May 2026high0.1.2.3
canada_exports_kcl_t22.9 milliontonnes KCl (38.7% of world exports); US 53%, Brazil 14%, China 6%2024Natural Resources Canada, Potash factshigh0.1.2.3
canada_exports_value_cad9 billionCAD per year2025Natural Resources Canada, reported by Fertilizer Daily, 30 Sep 2026medium0.1.2.3
export_routesRail from 10 Saskatchewan mines to Neptune Terminals (Vancouver); Canpotex also ships ~3 Mt/yr via Portland, Oregon (US soil); Nutrien selected Longview, WA for a new terminaltext2025-26Nasdaq/Investing News on Nutrien Longview; Railway Age, 'Potash is in the headlines', 2026medium0.1.2.3
price_shock_analogue1.78potash price increase Apr 2021-Apr 2022 (+178%) when Russia/Belarus supply was disrupted2022World Bank data via Union of Concerned Scientists, 2022medium0.1.2.3
demand_response_analogue-0.162change in US potash consumption 2020-21 as prices surged2021USDA ERS, ERR-354 (2023)high0.1.2.3
fertilizer_share_farm_costs0.15share of US farm cash costs (corn ~36-40% of operating costs)2022USDA AMS fertilizer comment summary; USDA ERS ERR-354; University of Illinois Extension 2025medium0.1.2.3
0.1.2.4OILOil+0.1+1.6

Oil

Crude reaches the US from whatever part of the basin it holds. Barrels that stop flowing leave the world market; spare capacity and a reserve release offset part of it; gasoline prices hit US opinion directly.

Switch oil.enabledVersions 0.1.2.4Parameters 15

What it is worth now

ScenarioOutcome countedPoints, module on against off95% interval
Full invasionAmericans go home+0.1 (within noise)−0.0 to +0.2
Alberta onlyAmericans go home+1.6+1.2 to +2.0
No invasionUS abandons the campaign+10.0+8.9 to +11.0

Everything else on. 3,000 paired runs, v0.1.3.8. Zero in the no-invasion scenario usually means the module has no path there.

Parameters and their sources (15)
ParameterValue or low – likely – highUnitSourceAdded in
oil.basin_share0.92share of Canadian crude from Alberta+Saskatchewan (scenario may override, e.g. 0.84 for Alberta only)CER 2025 provincial shares0.1.2.4
oil.occupied_output0.4 – 0.65 – 0.85share of basin output that keeps flowing under full US control (skilled workforce, non-cooperation, sabotage risk)Modeller estimatejudgement0.1.2.4
oil.offshore_diversion0 – 0.05 – 0.15share of US-bound crude Canada can re-route offshore if flows to the US stop (TMX already ~94% full)Modeller estimate from TMX utilization (RBN Energy, Aug 2026)judgement0.1.2.4
oil.price_per_mbd4 – 7 – 12USD/bbl world price rise per 1 Mb/d removed from the market (short run)Modeller estimate; short-run supply/demand elasticity literaturejudgement0.1.2.4
oil.market_adjust_max0.3 – 0.5 – 0.7share of the removed supply offset by spare capacity and demand response once markets adjustModeller estimatejudgement0.1.2.4
oil.market_adjust_months4 – 8 – 14months to reach full market adjustmentModeller estimatejudgement0.1.2.4
oil.spr_release_mbd0.5 – 1 – 1.5US Strategic Petroleum Reserve release rateModeller estimate; 2022 release ran ~1 Mb/d for six monthsjudgement0.1.2.4
oil.spr_months6 – 12 – 18months the SPR release can be sustainedModeller estimatejudgement0.1.2.4
oil.midwest_premium_cents10 – 20 – 35national-average gasoline cents/gal added by Midwest heavy-crude shortage at a full Canadian cut-offModeller estimate; PADD 2 runs 73% Canadian crude (Kansas City Fed)judgement0.1.2.4
oil.approval_per_10c0.003 – 0.006 – 0.009US support lost per 10-cent gasoline riseHarbridge, Krosnick & Wooldridge (2016)0.1.2.4
oil.approval_cap0.04 – 0.07 – 0.1maximum support drag from gasoline prices (saturating effect)Gupta, Pierdzioch & Tiwari (2025) find the penalty flattens past a threshold; cap is a modeller estimatejudgement0.1.2.4
oil.east_gdp_weight1.5 – 3 – 5% Canadian GDP lost at a total loss of eastern crude supply (Line 5 + US crude imports)Modeller estimatejudgement0.1.2.4
oil.east_dependence0.55 – 0.65 – 0.75share of Ontario/Quebec/NB refinery crude that arrives via US territory or from the USCAPP 2025; Line 5 filings (>50% Ontario, ~2/3 Quebec); NB Irving imports US crude0.1.2.4
oil.east_substitution_max0.3 – 0.5 – 0.7share of eastern gap covered by seaborne imports (Saint John, Quebec City, Montreal) once arrangedModeller estimate; contested if US navy interdicts the St. Lawrencejudgement0.1.2.4
oil.east_substitution_months1 – 3 – 6months to arrange seaborne replacementModeller estimatejudgement0.1.2.4
Data: data/commodities/oil.yaml (11 entries)
EntryValueUnitAs ofSourceConfidenceAdded in
canada_crude_to_us_mbd3.9Mb/d crude exported to US (90.1% of 4.3 Mb/d crude exports)2025Canada Energy Regulator, Market Snapshot: Overview of 2025 Canada-U.S. Energy Trade (27 May 2026); EIA Today in Energy 29 Jul 2026high0.1.2.4
canada_crude_to_us_value_cad126.1 billionCAD per year2025Canada Energy Regulator 2025 trade snapshot (C$140.0B total crude exports, 90.1% to US)high0.1.2.4
canada_share_us_crude_imports0.634share of US crude imports from Canada2025Canada Energy Regulator 2025 trade snapshothigh0.1.2.4
padd2_canadian_share_of_crude_inputs0.73share of Midwest refinery crude inputs that are Canadian (2.8 Mb/d); >70% of US imports of Canadian crude refined in Midwest2024Federal Reserve Bank of Kansas City, 'Canadian oil important for Midwest gasoline prices' (2025)high0.1.2.4
padd2_heavy_replacement_noteLandlocked Midwest refiners cannot easily replace Canadian heavy crude; Capline reversed in 2021 so Gulf heavy crude cannot move northtext2025S&P Global Commodity Insights, 31 Jan 2025high0.1.2.4
canada_production_mbd5.35Mb/d crude and equivalents; Alberta 83.8%, Saskatchewan 8.2%, NL 4.5%, BC 2.8%2025Canada Energy Regulator, Market Snapshot: record crude oil production in 2025 (17 Jun 2026)high0.1.2.4
tmx_capacity_mbd0.89Mb/d nominal; ~94% utilized Q2 2026; +90 kb/d expected end-20262026Trans Mountain fact sheet Jul 2026; RBN Energy Aug 2026high0.1.2.4
canada_non_us_crude_exports_mbd0.43Mb/d2025Asia Pacific Foundation of Canada via Gulf News 2026medium0.1.2.4
line5_eastern_dependenceLine 5 (Superior WI -> Sarnia, through Michigan) supplies >50% of Ontario refinery feedstock and is a critical source for ~2/3 of Quebec's crude via Line 9; all Enbridge routes to Sarnia cross US territorytext2025Canadian government filings reported by Western Standard (2025); Fraser Institute; CER Enbridge Mainline profilehigh0.1.2.4
canada_crude_imports_from_us_mbd0.375Mb/d US crude into Canada (NB 44%, QC 33%, ON 13%)2024CAPP, Canadian Imports of US Crude, Natural Gas and Refined Products (Jan 2025); CER 2025 snapshot (0.4 Mb/d)high0.1.2.4
gas_price_approval-0.006change in presidential approval per 10-cent rise in gasoline price (1976-2007)2016Harbridge, Krosnick & Wooldridge (2016), Political Psychology; nonlinear/saturating per Gupta, Pierdzioch & Tiwari (2025), American Politics Researchmedium0.1.2.4
0.1.2.5ELCElectricity00

Electricity

Cross-border power by region. Either side can cut it. The price effect is small; what matters is blackouts and who is blamed for them.

Switch electricity.enabledVersions 0.1.2.5Parameters 10

What it is worth now

ScenarioOutcome countedPoints, module on against off95% interval
Full invasionAmericans go home+0.0 (within noise)+0.0 to +0.0
Alberta onlyAmericans go home+0.0 (within noise)+0.0 to +0.0
No invasionUS abandons the campaign+0.2+0.1 to +0.4

Everything else on. 3,000 paired runs, v0.1.3.8. Zero in the no-invasion scenario usually means the module has no path there.

Parameters and their sources (10)
ParameterValue or low – likely – highUnitSourceAdded in
electricity.hydro_condition0.5 – 0.75 – 1.1Canadian hydro export capacity relative to a normal-water year (low runoff since 2023)Hydro-Quebec exports fell from 36 TWh (2021) to 15 TWh (2024); modeller estimatejudgement0.1.2.5
electricity.war_disruption_share0 – 0.2 – 0.4share of general trade disruption that applies to power flows absent a deliberate cut (grids are jointly operated)Modeller estimatejudgement0.1.2.5
electricity.us_support_drag0.002 – 0.006 – 0.012US support lost at a full cut of normal-water Canadian exports (regional bills; ~15% of US population in affected grids); level, not cumulativeModeller estimatejudgement0.1.2.5
electricity.us_gdp_weight0.01 – 0.03 – 0.06% US GDP lost at a full cut of normal-water Canadian exportsModeller estimate; Canada supplies 32.7 TWh vs ~4,200 TWh US consumptionjudgement0.1.2.5
electricity.blackout_prob0.03 – 0.1 – 0.25monthly chance of attributable load-shedding in a US import region during peak months (Jun-Aug, Jan) under a full cutModeller estimate; NYISO and ISO-NE margins tight in heat waves and cold snapsjudgement0.1.2.5
electricity.blackout_support_effect-0.03 – 0.01 – 0.04lasting change in US support after a blackout blamed on Canada (positive = rally behind the war)Modeller estimate; Ontario's 2025 surcharge drew escalation, not concessionjudgement0.1.2.5
electricity.occupied_output0.3 – 0.6 – 0.85share of exports that keep flowing when the US controls the exporting province (remote northern dams and long lines)Modeller estimatejudgement0.1.2.5
electricity.can_gdp_weight0.1 – 0.3 – 0.6% Canadian GDP lost when US stops exports to Canada, at normal water (rises in drought)Modeller estimate; BC imported 25% of its needs in fiscal 2024judgement0.1.2.5
electricity.can_blackout_prob0.03 – 0.08 – 0.2monthly chance of Canadian load-shedding in winter (Dec-Feb) when US imports are cutModeller estimate; Alberta came within ~200 MW of rolling blackouts in Jan 2024judgement0.1.2.5
electricity.can_blackout_gdp0.3 – 1 – 2% Canadian GDP lost in a month with load-sheddingModeller estimatejudgement0.1.2.5
Data: data/commodities/electricity.yaml (10 entries)
EntryValueUnitAs ofSourceConfidenceAdded in
canada_exports_twh32.7TWh exported to US (35.7 in 2024, 49.4 in 2023)2025Canada Energy Regulator, Overview of 2025 Canada-U.S. Energy Trade (27 May 2026); CER 2024 and 2023 snapshotshigh0.1.2.5
canada_exports_value_cad3.3 billionCAD per year (0.4% of Canadian goods exports)2025Canada Energy Regulator 2025 trade snapshothigh0.1.2.5
canada_imports_twh22.1TWh imported from US, C$1.4B2025Canada Energy Regulator 2025 trade snapshothigh0.1.2.5
canada_share_us_electricity_imports0.813share of US electricity imports supplied by Canada2025Canada Energy Regulator 2025 trade snapshothigh0.1.2.5
hq_exports_history_twh36.2 (2021), 35.6 (2022), 23.0 (2023), 15.1 (2024); low runoff since 2023text2024Hydro-Quebec annual report 2024 via Energi Mediahigh0.1.2.5
chpe1,250MW HVDC Quebec -> New York City, commercial operation May 2026; up to ~20% of NYC demand; outages June-July 20262026S&P Global 13 May 2026; EIA Today in Energy 20 Jul 2026high0.1.2.5
necec_and_new_england0.13net imports share of New England supply, spring 2026 (7% in spring 2025); NECEC delivery obligation 9.55 TWh/yr; NECEC curtailed during Winter Storm Fern (Jan 2026) to meet Quebec demand2026ISO-NE Internal Market Monitor Spring 2026 report via Acadia Center (Oct 2026)high0.1.2.5
ontario_us_customers1.5 millionUS homes and businesses in NY, MI, MN served by Ontario exports2025Ontario government, March 2025 (CP24, SupplyChainBrain)medium0.1.2.5
ontario_surcharge_analogue25% surcharge on Ontario exports to NY, MI, MN took effect 10 Mar 2025 and was suspended 11 Mar 2025 after the US threatened to double steel and aluminum tariffs and declare a 'National Emergency on Electricity'text2025Blakes; CBC; Utility Dive (March 2025)high0.1.2.5
bc_import_dependence0.25share of BC consumption imported in fiscal 2024 (13.6 TWh, record; drought)2024Electricity Forum; Global News (2024-25)medium0.1.2.5
0.1.2.7LOGSupply and logistics−0.5+0.1

Supply and logistics

Sustainment for both armies and for civilians. Interdiction of thin east-west corridors holds occupier strength down and costs Canadians too. Canada's munitions run short early.

Switch logistics.enabledVersions 0.1.2.7Parameters 16

What it is worth now

ScenarioOutcome countedPoints, module on against off95% interval
Full invasionAmericans go home−0.5−0.9 to −0.2
Alberta onlyAmericans go home+0.1 (within noise)−0.4 to +0.7
No invasionUS abandons the campaign+0.0 (within noise)−0.2 to +0.3

Everything else on. 3,000 paired runs, v0.1.3.8. Zero in the no-invasion scenario usually means the module has no path there.

Parameters and their sources (16)
ParameterValue or low – likely – highUnitSourceAdded in
logistics.interdiction_coef0.05 – 0.12 – 0.25share of occupier supply flow interdicted per unit of resistance index R (sabotage and blockade of road/rail, aggregate)Modeller estimate; 2020 blockades showed a few sites can halt a quarter of CN's networkjudgement0.1.2.7
logistics.network_fragility1.2 – 1.5 – 2multiplier on interdiction for Canada's thin east-west corridors (single rail and road links around Lake Superior and through eastern Ontario)Modeller estimate; Transport Canada 2020 (CN's only eastern link near Belleville)judgement0.1.2.7
logistics.winter_logistics_mult0.5 – 1 – 1.5added interdiction effect per unit of blizzard/freezing-rain day share (complex weather)Modeller estimatejudgement0.1.2.7
logistics.shortfall_strength_sensitivity0.3 – 0.5 – 0.8loss of effective occupier strength per unit supply shortfall (stocks and rerouting buffer the rest)Modeller estimatejudgement0.1.2.7
logistics.interdiction_cap0.35maximum monthly supply shortfallModeller estimatejudgement0.1.2.7
logistics.rail_sabotage_can_gdp0.1 – 0.25 – 0.5% Canadian GDP lost per month at full interdiction (the same lines carry Canadian goods)2020 blockades: ~0.25% of GDP for the month (RBC via BNN Bloomberg)0.1.2.7
logistics.caf_days_of_supply3 – 7 – 15days of high-intensity combat before key munitions run shortOAG Oct 2026 (significant stockpile gaps); former CDS Eyre: 'out in days in some cases'; NATO target 30 days0.1.2.7
logistics.caf_ammo_penalty0.3 – 0.5 – 0.7loss of Canadian conventional effectiveness once key munitions are exhaustedModeller estimatejudgement0.1.2.7
logistics.allied_resupply_relief0.3 – 0.5 – 0.7share of the munitions penalty removed when allies supply materiel (external level >= materiel)Modeller estimatejudgement0.1.2.7
logistics.can_food_pain0.3 – 0.8 – 1.5% GDP-equivalent pain for Canada at a full cut of US food imports in winter (half in summer)Modeller estimate; US supplies 56.5% of vegetable and ~36% of fruit imports (AAFC 2025; UBC)judgement0.1.2.7
logistics.can_pharma_pain0.2 – 0.5 – 1% GDP-equivalent pain for Canada at a full cut of US pharmaceutical importsModeller estimate; US 30.8% of pharma imports (StatCan 2024)judgement0.1.2.7
logistics.substitution_max0.4 – 0.6 – 0.8share of lost US food and drug supply Canada replaces via Mexico, Europe and its own portsModeller estimatejudgement0.1.2.7
logistics.substitution_months2 – 4 – 6months to reach maximum substitutionModeller estimatejudgement0.1.2.7
logistics.jit_shock_us0.1 – 0.3 – 0.6% US GDP lost in the first month of a trade cut from just-in-time chains (autos, parts), decayingModeller estimate; 2022 Ambassador Bridge blockade shut US and Canadian plants within daysjudgement0.1.2.7
logistics.jit_shock_can1 – 2 – 3% Canadian GDP lost in the first month of a trade cut from just-in-time chains, decayingModeller estimatejudgement0.1.2.7
logistics.jit_halflife_months2 – 3 – 5half-life of the just-in-time shock as supply chains rerouteModeller estimatejudgement0.1.2.7
Data: data/commodities/logistics.yaml (8 entries)
EntryValueUnitAs ofSourceConfidenceAdded in
ambassador_bridge25-30% of Canada-US trade, ~US$323-360M a day; Feb 2022 week-long blockade shut or slowed Ford, Toyota, Honda and GM plants on both sides within days; ~US$100-141M a day of auto tradetext2022The Conversation / U of Guelph (Feb 2022); Fortune (14 Feb 2022, Reuters, IHS Markit); CBC (Feb 2022)high0.1.2.7
canada_veg_imports_us_share0.565US share of Canadian fresh/frozen field vegetable imports (C$2.31B of C$3.90B); 59.5% in 20242025Agriculture and Agri-Food Canada, Statistical Overview of the Canadian Field Vegetable Industry 2025high0.1.2.7
canada_fruit_imports_us_share0.36US share of Canadian fruit imports (vegetables 67%), 2010-2022 average2022UBC Canada Food Flows via CBC (2025)high0.1.2.7
canada_produce_import_dependenceCanada imports 88% of fresh fruit and nuts and 72% of vegetables consumed; produce ~15% of grocery spendingtext2026Federal food blueprint via The Packer (2026)medium0.1.2.7
canada_pharma_imports_us_share0.308US share of Canadian pharmaceutical imports (Europe 33.5%, India 3.2%, China 2.9%)2024Statistics Canada via CTV/CP24 (July 2025)high0.1.2.7
caf_ammunitionOAG (Oct 2026): strategic stockpile has 'significant gaps' against requirements that will take years to fix; C$9.2B over 13 years committed in 2024, no stockpile deliveries yet by Mar 2026. Former CDS Eyre: at Ukraine-like consumption 'we would be out in days in some cases'; NATO asks for 30 days of supplytext2026Auditor General of Canada, Report 24 (Oct 2026); CTV (2024); CBC (2023)high0.1.2.7
f35_dependenceMajor maintenance, overhaul and software upgrades on the F-35 happen in the US; no confirmed kill switch, but the US controls software, upgrades, parts and armamenttext2025CBC (2025, Minister Miller); 19FortyFive (2025); DND spokesperson via scribe (2025)medium0.1.2.7
rail_blockade_2020Feb 2020 blockades shut CN's eastern network (~1/4 of operations), cancelled 400+ trains, halted perishable food, chlorine and propane by rail; Quebec rationed propane with ~1 week of inventory; GDP hit pencilled at ~0.25% for the monthtext2020Transport Canada, Transportation in Canada 2020; CBC; BNN Bloomberg (RBC estimate); Energy Nowhigh0.1.2.7

Allies and third parties

0.1.2.8NATONATO and allies+5.3+9.0

NATO and allies

NATO as an institution is blocked by the US veto, so this is a coalition of the willing. Each ally is drawn separately, in correlated blocs, and lands on words, sanctions and basing denial, materiel, or forces.

Switch nato.enabledVersions 0.1.2.8, 0.1.2.11Parameters 35

What it is worth now

ScenarioOutcome countedPoints, module on against off95% interval
Full invasionAmericans go home+5.3+4.3 to +6.2
Alberta onlyAmericans go home+9.0+7.9 to +10.2
No invasionUS abandons the campaign−1.0−1.7 to −0.4

Everything else on. 3,000 paired runs, v0.1.3.8. Zero in the no-invasion scenario usually means the module has no path there.

Parameters and their sources (35)
ParameterValue or low – likely – highUnitSourceAdded in
nato.ceiling_weightsdiplomatic 0.25, sanctions 0.4, materiel 0.3, forces 0.05highest level of allied response over the war (NATO as an institution is blocked by US veto; this is a coalition of the willing)Modeller estimate; Greenland 2026 shows allies will deploy symbolic forces before a war, not that they would fight the USjudgement0.1.2.8
nato.tripwire_escalation0.1 – 0.25 – 0.4probability mass moved up one or two levels when allied troops are in Canada and the US attacks themModeller estimatejudgement0.1.2.8
nato.us_counter_leverage0.1 – 0.2 – 0.35probability mass moved down when the US threatens to abandon European defence (Europe relies on US enablers against Russia)Modeller estimate; ~80,000 US troops in Europe (AP 2026)judgement0.1.2.8
nato.sanctions_month1 – 2 – 4war month in which coalition sanctions take effectModeller estimate; EU sanctions on Russia took days to weeks in 2022judgement0.1.2.8
nato.materiel_month2 – 4 – 8war month in which materiel support reaches CanadaModeller estimatejudgement0.1.2.8
nato.forces_month2 – 5 – 10war month in which allied forces enter (if the ceiling is forces)Modeller estimatejudgement0.1.2.8
nato.tripwire_casualty_penalty0.01 – 0.03 – 0.06lasting drop in US support when US forces kill allied soldiers in CanadaModeller estimatejudgement0.1.2.8
nato.basing_loss_drag0.005 – 0.015 – 0.03lasting drop in US support once allies restrict US basing and access (loss of global reach; elite and public reaction)Modeller estimatejudgement0.1.2.8
nato.norad_break_posture0.02 – 0.05 – 0.1loss of Canadian defensive posture when NORAD's shared air picture is cut at the outsetModeller estimate; NORAD is a binational commandjudgement0.1.2.8
nato_members.band_position-1 – 0 – 1where every ally's probabilities sit inside the report's low-high band this run (-1 low, +1 high); shared uncertainty about absolute levelsDeep-research report (12 Nov 2026) s7.2 bands; shared position is a modelling choice0.1.2.11
nato_members.common_corr0.15 – 0.25 – 0.4correlation between any two allies' responses from a common 'European resolve' factor (zero for the US-leaning bloc)Report s7.4 recommends a common factor but gives no number; modeller estimatejudgement0.1.2.11
nato_members.delay_jitter_sigma0.4lognormal sigma on each ally's delay to each tierModeller estimatejudgement0.1.2.11
nato_members.t4_delay_months0.25 – 0.5 – 0.75months for a JEF-style or ad hoc deploymentReport s4.2: 1-3 weeks; Greenland Jan 2026 took about 10 days0.1.2.11
nato_members.eu_t2_delay_months1 – 2 – 6months for EU trade countermeasuresReport s4.2: weeks for rebalancing, about six months for the Anti-Coercion Instrument0.1.2.11
nato_members.eu_targeted_s1_t20.4chance of EU trade measures under economic coercion when the EU itself is targetedReport s7.2 and s7.60.1.2.11
nato_members.germany_eu_halving0.5multiplier on EU collective action when Germany does not reach sanctionsReport s7.40.1.2.11
nato_members.us_counter_chill0.05 – 0.075 – 0.1drop in second movers' probabilities (blocs B and C, sanctions and above) under US counter-movesReport s7.4: 5-10pp; Iran-war precedent (5,000 troops ordered out of Germany, floated suspension of Spain)0.1.2.11
nato_members.abandon_frontline_shift0.05 – 0.1 – 0.2drop in frontline states' probabilities (sanctions and above) when the US threatens to abandon Europe's defenceDirection from report s7.4 (frontline states hedge toward Washington); size is a modeller estimatejudgement0.1.2.11
nato_members.abandon_west_shift0 – 0.05 – 0.1rise in western core states' condemnation and sanctions probabilities under the same threatDirection from report s7.4 (western states harden); size is a modeller estimatejudgement0.1.2.11
nato_members.russia_concurrent_prob0.05 – 0.15 – 0.3chance Russia escalates in Europe during the crisis (used when the second-war module is off)Modeller estimatejudgement0.1.2.11
nato_members.russia_frontline_shift0.2 – 0.3 – 0.4drop in frontline states' materiel and forces probabilities if Russia movesReport s7.4: -20 to -40pp0.1.2.11
nato_members.russia_other_shift0.1same, for other alliesReport s7.40.1.2.11
nato_members.uk_t3_cascade0.1rise in the core bloc's materiel probability when the UK sends materielReport s7.40.1.2.11
nato_members.poland_cascade0.05 – 0.15 – 0.25rise in the Baltic states' and Romania's sanctions probability when Poland goes beyond wordsDirection from report s7.4; size is a modeller estimatejudgement0.1.2.11
nato_members.killed_allies_t2_shift0.15 – 0.2 – 0.25rise in sanctions probability across blocs A and B when US forces kill allied personnelReport s7.4: +15-25pp0.1.2.11
nato_members.killed_allies_t4_shift0.1rise in deployment probability across blocs A and B in the same caseReport s7.40.1.2.11
nato_members.combat_share_bloc_a0.1 – 0.18 – 0.25share of core-bloc deployers that would fight US forces (zero for other blocs)Report s7.2 and s8: direct combat is about 10-25% of deployments for Bloc A0.1.2.11
nato_members.combat_saturation_share0.25share of non-US NATO GDP among combatants at which the forces strength reaches 1Modeller estimate (about France plus the UK)judgement0.1.2.11
nato_members.materiel_saturation_share0.15 – 0.25 – 0.4share of non-US NATO GDP among donors at which materiel strength reaches its ceilingModeller estimatejudgement0.1.2.11
nato_members.delivery_feasibility_s30.3 – 0.4 – 0.5share of materiel that reaches an invaded Canada (US controls the sea and air approaches)Report s8: about 0.3-0.5 in a full invasion0.1.2.11
nato_members.delivery_feasibility_s20.5 – 0.7 – 0.9same, in a limited seizureModeller estimatejudgement0.1.2.11
nato_members.sanctions_reference0.6 – 0.75 – 0.9combined basing-and-economic weight at which sanctions strength reaches 1 (the report's central split holds about 75% of non-US NATO GDP and 77% of US troops in Europe)Report s7.5; mapping to model strength is a modelling choice0.1.2.11
nato_members.reform_uk_mult0.5multiplier on all UK probabilities under a Reform-led governmentReport s7.60.1.2.11
nato_members.rn_france_mult0.5multiplier on France's probabilities under an RN presidencyModeller estimate; report s7.6 says an RN win removes the lead moverjudgement0.1.2.11
nato_members.rn_bloc_a_shift0.1 – 0.125 – 0.15drop in the rest of the core bloc's condemnation and sanctions probabilities under an RN presidencyReport s7.6: 10-15pp0.1.2.11
Data: data/commodities/nato.yaml (4 entries)
EntryValueUnitAs ofSourceConfidenceAdded in
article5Article 5 obliges each ally to assist an attacked ally but leaves each free to decide what action to take; the North Atlantic Council decides by consensus, so any member, including the attacker, can block a collective decision. The treaty has no procedure for one member attacking another and no expulsion mechanismtext2026NATO, 'Collective defence and Article 5'; Belfer Center explainer; Law & More treaty summaryhigh0.1.2.8
greenland_precedentJan 2026 (Operation Arctic Endurance): France, Germany, UK, Norway, Sweden, Finland and the Netherlands sent small contingents to Greenland (France 15, Germany 13; Denmark ~350) as a tripwire against US pressure; the US answered with 10% tariffs on eight allies and said the troops would not affect its decisiontext2026AP (15 Jan 2026); Euromaidan Press (15 and 20 Jan 2026); The Aviationist (16 Jan 2026)high0.1.2.8
us_forces_europe80,000US troops in Europe (80,000-100,000; >36,000 Germany, >12,000 Italy, ~10,000 UK); NDAA floor 76,000; drawdown under way in 20262026AP via WTOP (May 2026); Air & Space Forces Magazine (June 2026)high0.1.2.8
europe_dependence_noteUS presence in Europe also projects US power to Africa and the Middle East (EUCOM commander, March 2026); Europe still relies on US enablers against Russiatext2026AP via KTAR (May 2026)medium0.1.2.8
Data: data/alliances/nato_members.yaml (1 entries)
EntryValueUnitAs ofSourceConfidenceAdded in
_metatable: tiers, us_troops, gdp, frontline, delays, bands, group_correlation, caveatsDeep-research report 'How NATO Allies Would Respond if the US Coerced or Attacked Canada' (12 Nov 2026), sections 3.1, 7.1, 7.2 and 7.4. Tier probabilities, blocs, delays and correlations are that report's ANALYTICAL ESTIMATES, not forecasts; it notes they encode ordering and correlation more firmly than absolute levels.0.1.2.10
0.1.2.8.1ARMForeign small arms0−1.2

Foreign small arms

Arms and ammunition for the resistance from domestic stock, the cross-border black market and state suppliers. Aggregate flows only.

Switch arms.enabledVersions 0.1.2.8.1Parameters 10

What it is worth now

ScenarioOutcome countedPoints, module on against off95% interval
Full invasionAmericans go home+0.0 (within noise)−0.1 to +0.1
Alberta onlyAmericans go home−1.2−1.6 to −0.8
No invasionUS abandons the campaign+0.0 (within noise)+0.0 to +0.0

Everything else on. 3,000 paired runs, v0.1.3.8. Zero in the no-invasion scenario usually means the module has no path there.

Parameters and their sources (10)
ParameterValue or low – likely – highUnitSourceAdded in
arms.domestic_stock_months6 – 12 – 24months of resistance ammunition from Canadian civilian and cached military stocks at full tempoModeller estimate; 2.47M licence holders (RCMP 2025); CAF war stocks have significant gaps (OAG 2026)judgement0.1.2.8
arms.ammo_floor0.3 – 0.5 – 0.7resistance lethality multiplier once domestic ammunition is exhausted and nothing arrivesModeller estimatejudgement0.1.2.8
arms.black_market_supply0.2 – 0.4 – 0.6share of resistance ammunition and small-arms need the US-origin black market can meet (profit-driven)Modeller estimate; US-origin guns dominate Ontario crime guns (91% of handguns, 2024) but only 20% of nationally traced crime guns (RCMP 2023)judgement0.1.2.8
arms.black_market_interdiction0.1 – 0.2 – 0.35share of black-market flow stopped (long land border; US enforcement on its own side in wartime)Modeller estimatejudgement0.1.2.8
arms.state_supply0.4 – 0.7 – 1share of need met by foreign state shipments when openModeller estimatejudgement0.1.2.8
arms.state_interdiction_normal0.2 – 0.35 – 0.5share of state shipments intercepted without a declared blockadeModeller estimatejudgement0.1.2.8
arms.state_interdiction_blockade0.5 – 0.7 – 0.85share of state shipments intercepted under a US air and sea blockade of CanadaModeller estimatejudgement0.1.2.8
arms.quality_gain0.1 – 0.25 – 0.4added resistance lethality at full supply of military-grade small arms and light weapons (vs hunting rifles and shotguns)Modeller estimatejudgement0.1.2.8
arms.exposure_prob0.02 – 0.05 – 0.1monthly chance a state shipment is publicly exposedModeller estimatejudgement0.1.2.8
arms.exposure_us_rally-0.01 – 0.005 – 0.02lasting rise in US support after an exposure ('foreign powers arming insurgents'); negative = no rally or backfireModeller estimate; belongs in the narrative module (0.1.21b)judgement0.1.2.8
Data: data/commodities/arms.yaml (3 entries)
EntryValueUnitAs ofSourceConfidenceAdded in
canada_licence_holders2,473,661firearms licence holders (1,626,754 non-restricted; 794,768 restricted privileges), ~6% of population2025RCMP, 2025 Commissioner of Firearms Reporthigh0.1.2.8
canada_registered_restricted1,269,076registered restricted and prohibited firearms (long guns are not registered)2024RCMP, 2024 Commissioner of Firearms Reporthigh0.1.2.8
us_origin_crime_gunsOntario: 91% of crime handguns recovered in 2024 came from the US (Ontario government via NYT, 2025); Toronto: 88% of crime guns smuggled in 2024, up from 51% in 2014. National RCMP tracing 2023: 80% of traced crime guns with known source were domestic, 20% smuggled. A US$500 handgun sells for up to C$5,000 in Canadatext2024New York Times (Dec 2025); PolySeSouvient summary of RCMP CNFTC data (June 2026); CBC (2025)medium0.1.2.8
0.1.2.12AIRAllied air cover+0.8+2.1

Allied air cover

Allies will not fight on the ground but offer to fly a combat air patrol over Canada. The US respects it or shoots.

Switch allied_air.enabledVersions 0.1.2.12Parameters 14

What it is worth now

ScenarioOutcome countedPoints, module on against off95% interval
Full invasionAmericans go home+0.8+0.5 to +1.2
Alberta onlyAmericans go home+2.1+1.5 to +2.6
No invasionUS abandons the campaign+0.0 (within noise)+0.0 to +0.0

Everything else on. 3,000 paired runs, v0.1.3.8. Zero in the no-invasion scenario usually means the module has no path there.

Parameters and their sources (14)
ParameterValue or low – likely – highUnitSourceAdded in
allied_air.offer_given_materiel0.2 – 0.4 – 0.6chance allies offer to fly a combat air patrol over Canada when their response reaches materiel but not ground forces (used when offer is 'auto')Modeller estimatejudgement0.1.2.12
allied_air.coverage0.1 – 0.25 – 0.45share of the theatre and clock an allied air patrol can cover with warning (continental frontage, few tankers, long transit)Modeller estimate; air refuelling and enablers are a known European shortfall, and much of NATO's enabling structure runs on US systems (deep-research report s4.1)judgement0.1.2.12
allied_air.no_warning_mult0.1 – 0.3 – 0.5multiplier on coverage when the patrol has to deploy after the attack startsModeller estimatejudgement0.1.2.12
allied_air.us_platform_penalty0.15 – 0.3 – 0.45loss of coverage because part of the allied fighter fleet depends on US-controlled support (F-35 software, spares, logistics)Modeller estimate; 12 European states had 466 F-35s on order or preselected (SIPRI, March 2026, via the deep-research report)judgement0.1.2.12
allied_air.us_engages_prob0.25 – 0.5 – 0.75chance the US attacks the allied patrol instead of working around itModeller estimate; in the Greenland crisis the US backed down rather than push allied troops aside, but nothing was at stake militarilyjudgement0.1.2.12
allied_air.denial_effect0.5 – 0.75 – 1share of the US air-superiority advantage removed where and while a respected patrol fliesModeller estimatejudgement0.1.2.12
allied_air.occupation_effect0.05 – 0.15 – 0.3loss of occupier effectiveness at full coverage (less air support and surveillance)Modeller estimatejudgement0.1.2.12
allied_air.capitulation_damping0.1 – 0.3 – 0.5cut in Canada's capitulation hazard at full coverage (a government and territory under allied air cover)Modeller estimatejudgement0.1.2.12
allied_air.delivery_gain0.2 – 0.5 – 1rise in materiel delivery at full coverage (an air bridge stays open)Modeller estimatejudgement0.1.2.12
allied_air.survival_months0.1 – 0.3 – 1months an engaged patrol lasts against US air powerModeller estimatejudgement0.1.2.12
allied_air.shootdown_support_penalty0.02 – 0.05 – 0.09lasting drop in US support when US forces shoot down allied aircraftModeller estimate; larger than the ground tripwire penalty because the US chose to fire first on allies in the airjudgement0.1.2.12
allied_air.escalation_prob0.05 – 0.15 – 0.3chance that allies who lose aircraft enter the war instead of standing downModeller estimatejudgement0.1.2.12
allied_air.escalation_forces0.2 – 0.4 – 0.7allied forces strength (0-1) if they enter the warModeller estimatejudgement0.1.2.12
allied_air.war_risk_mult1.2 – 1.5 – 2multiplier on the US withdrawal hazard while allied aircraft are being engaged or after allies enter (risk of a wider war)Modeller estimatejudgement0.1.2.12
0.1.2.132WA second war+2.1+1.5

A second war

While the US is committed in Canada, another state attacks a US ally. The US defends the ally or abandons it.

Switch second_war.enabledVersions 0.1.2.13Parameters 16

What it is worth now

ScenarioOutcome countedPoints, module on against off95% interval
Full invasionAmericans go home+2.1+1.6 to +2.6
Alberta onlyAmericans go home+1.5+1.0 to +2.0
No invasionUS abandons the campaign+3.2+2.5 to +3.8

Everything else on. 3,000 paired runs, v0.1.3.8. Zero in the no-invasion scenario usually means the module has no path there.

Parameters and their sources (16)
ParameterValue or low – likely – highUnitSourceAdded in
second_war.base_hazard_monthly0.001 – 0.003 – 0.008monthly chance that some adversary attacks a US ally or partner, before opportunismModeller estimatejudgement0.1.2.12
second_war.tied_down_mult0.5 – 1.5 – 3rise in that hazard at full US commitment in Canada (share of sustainable ground forces deployed)Modeller estimatejudgement0.1.2.12
second_war.alliance_break_mult0.5 – 1.5 – 3rise in that hazard when allies have fully broken with the US (sanctions strength 1)Modeller estimatejudgement0.1.2.12
second_war.theatre_weightseurope 0.35, pacific 0.35, korea 0.1, middle_east 0.2where the second war startsModeller estimatejudgement0.1.2.12
second_war.defend_prob_base0.4 – 0.65 – 0.85chance the US defends the attacked ally when the alliance is intactModeller estimate; ECFR (June 2026): majorities in all 15 countries polled doubt the US would come to their aidjudgement0.1.2.12
second_war.defend_integrity_power1exponent on alliance integrity (1 - share of that region's allies that have broken with the US)Modelling choice0.1.2.12
second_war.ground_demand{'europe': 0.3, 'pacific': 0.12, 'korea': 0.35, 'middle_east': 0.15}share of sustainable US ground forces a defended second war pulls away from CanadaModeller estimatejudgement0.1.2.12
second_war.air_demand{'europe': 0.35, 'pacific': 0.55, 'korea': 0.4, 'middle_east': 0.3}share of US air power pulled awayModeller estimatejudgement0.1.2.12
second_war.us_gdp_hit{'europe': 1.0, 'pacific': 5.0, 'korea': 2.0, 'middle_east': 1.0}% US GDP lost per year while the second war lasts, if the US fights (half if it stays out)Modeller estimate; Bloomberg Economics (2024) put a Taiwan war at about 10% of world GDP, more for the economies involvedjudgement0.1.2.12
second_war.air_strength_weight0.1 – 0.2 – 0.35loss of occupier effectiveness in Canada when all US air power is pulled awayModeller estimatejudgement0.1.2.12
second_war.two_war_withdrawal_mult1.5 – 2.5 – 4multiplier on the US withdrawal hazard from Canada while it fights a second warModeller estimatejudgement0.1.2.12
second_war.rally_effect-0.03 – 0.02 – 0.06lasting change in US support for the Canada war when the US enters a second war (positive = rally)Modeller estimate; uncertain signjudgement0.1.2.12
second_war.abandon_support_drop0 – 0.02 – 0.05lasting drop in US support when the US leaves an attacked ally undefendedModeller estimatejudgement0.1.2.12
second_war.europe_aid_suppression0.5 – 0.75 – 0.95cut in European materiel and forces for Canada once Europe is itself at warDeep-research report s7.4: -20 to -40pp on frontline states' materiel and forces, -10pp elsewhere; converted to a cut in strength (modeller estimate)judgement0.1.2.12
second_war.europe_reconcile_prob0.2 – 0.5 – 0.8chance that Europeans drop sanctions and basing limits on the US when the US defends themModeller estimatejudgement0.1.2.12
second_war.duration_months3 – 12 – 36length of the second warModeller estimatejudgement0.1.2.12

Geography

0.1.3.0PRVProvinces+1.0+0.2

Provinces

Ten provinces and the North as separate areas of occupation, with real populations, rural dispersion, regional weather, provincial politics, resources by province, troop allocation rules and province-by-province pacification.

Switch spatial.enabledVersions 0.1.3.0, 0.1.3.1, 0.1.3.2, 0.1.3.2.1, 0.1.3.3, 0.1.3.4, 0.1.3.5Parameters 17

What it is worth now

ScenarioOutcome countedPoints, module on against off95% interval
Full invasionAmericans go home+1.0+0.6 to +1.5
Alberta onlyAmericans go home+0.2+0.1 to +0.4
No invasionUS abandons the campaign+0.0 (within noise)+0.0 to +0.0

Everything else on. 3,000 paired runs, v0.1.3.8. Zero in the no-invasion scenario usually means the module has no path there.

Parameters and their sources (17)
ParameterValue or low – likely – highUnitSourceAdded in
spatial.rural_dispersion_coef0 – 0.5 – 1extra troops per person needed as the rural share of a province rises (requirement x [1 + coef x (rural share - national rural share)])Modeller estimate. Force-ratio studies give troops per 1,000 people without a settlement term; the low end means dispersion does not matterjudgement0.1.3.1
spatial.presence_per_100k_km2100 – 300 – 800troops needed per 100,000 km2 of land simply to be present, whatever the populationModeller estimate; no historical anchor for holding very large, nearly empty territory. Uses total land area, which overstates the need where land is uninhabitedjudgement0.1.3.1
spatial.local_shock_share0.3 – 0.5 – 0.7share of the month-to-month variation in occupier casualties that is local to a province rather than shared nationallyModeller estimatejudgement0.1.3.1
spatial.compression_opposition0.3 – 0.5 – 0.7share of each province's peacetime gap from the national level in opposition that survives an overt invasion (1 = gaps unchanged, 0 = gaps vanish)Research report, 13 Nov 2026: after Russia's 2022 invasion regional gaps in Ukraine largely closed (KIIS: 83% east, 85% south opposed concessions; civic identity 77-82% in every region); Canada's 2025 tariff shock moved opinion the same way. Analytical estimatejudgement0.1.3.2
spatial.compression_armed0.5 – 0.7 – 0.9share of each province's gap in armed propensity that survives an overt invasionResearch report, 13 Nov 2026: capacity proxies (firearms, rural skills) do not converge the way attitudes do. Analytical estimatejudgement0.1.3.2
spatial.pacified_residual0.08 – 0.15 – 0.2share of a province's resistance that remains once it is pacified (it then needs only a holding garrison)Research report, 17 Nov 2026: residual violence in pacified regions was 8-20% of peak (Anbar 12-17%, Ramadi 16%, Malaya 13%, Basra 8%). Was 0.05/0.15/0.30 (modeller estimate) before v0.1.3.5judgement0.1.3.4
spatial.consolidation_share0.9share of the occupied population that must live in pacified provinces for the occupation to count as consolidatedModeller choice, so that one small province holding out does not block the outcomejudgement0.1.3.4
spatial.pacify_median_months12 – 24 – 48median months for a contested province to go quiet once the delay has passed, with the occupier at full strength there and moderate resistanceResearch report: contested regions that went quiet took a median of 50 months from the start of the conflict (n=23), but about 3-19 months from when the occupier had enough force and local partners. Recommended 12 / 24 / 480.1.3.5
spatial.pacify_delay_months12 – 15 – 18months after the occupier arrives in a province before it can go quiet at allResearch report: no contested region in the case table went quiet before month 24; recommends a near-zero hazard for the first 12-18 months0.1.3.5
spatial.pacify_control_exponent0.5 – 1 – 2how strongly the chance of going quiet depends on the occupier's strength in the province (2 = the national formula; lower = weaker)Modeller estimate, informed by the research report: troop density is only weakly linked to outcome (Friedman 2011: raising density from 5 to 80 per 1,000 adds under 15 points to success; Connable and Libicki: ratios 'neither accurate nor predictive')judgement0.1.3.5
spatial.hard_province_share0.2 – 0.275 – 0.35chance that a contested province is one of the hard ones that rarely go quietResearch report: about 25-30% of contested regions never went quiet while the conflict lasted (Ninewa, Diyala, Salah ad-Din, Helmand, Kandahar, Belfast, Derry, South Armagh)0.1.3.5
spatial.hard_median_months60 – 78 – 96median months to quiet for a hard province, at full strengthResearch report: hostile, identity-mobilized regions, median 60-96 months or more0.1.3.5
spatial.late_drop_month78months after arrival beyond which a province that is still contested becomes much less likely to go quietResearch report: regions not quiet by about year 6-7 almost never went quiet0.1.3.5
spatial.late_drop_mult0.1multiplier on the chance of going quiet after late_drop_monthModeller choice implementing the research finding abovejudgement0.1.3.5
spatial.relapse_base_monthly0.001 – 0.002 – 0.003monthly chance a pacified province turns violent again while its garrison is kept up and there is no outside supportResearch report: 0 of about 48 Malayan White Areas relapsed in 1953-60; about 13% of early Philippine civil provinces did, within months. Recommended 0.1-0.3% a month0.1.3.5
spatial.relapse_shortfall_monthly0.02 – 0.03 – 0.04added monthly chance of relapse when the garrison is entirely absent (scaled by the shortfall)Research report: Iraqi Sunni provinces relapsed about 2-2.5 years after the US left; 2-4% a month0.1.3.5
spatial.relapse_outside_monthly0.01added monthly chance of relapse at full outside support for the resistance (scaled by allied materiel and forces)Research report: the Afghan north relapsed after 7-8 years as outside pressure grew; about 1% a month0.1.3.5
Data: data/geography/canada_provinces.yaml (1 entries)
EntryValueUnitAs ofSourceConfidenceAdded in
canada_provincestable: units, population, land_area_km2, rural_population_2021, rural_share_2021, weather_regions2026-07-01Statistics Canada, Table 17-10-0009-01, population estimates for July 1, 2026 (preliminary), released 23 September 2026; provincial figures confirmed against the Manitoba Bureau of Statistics quarterly report of the same date; territorial figures as reported from The Daily of 23 September 2026. The 13 figures sum to the published national total of 41,798,407.0.1.3.0
Data: data/geography/political_geography.yaml (1 entries)
EntryValueUnitAs ofSourceConfidenceAdded in
alberta_referendum_2026table: remain, start_separation_process, turnout, dateElections Alberta official results, 29 October 2026, as reported by CBC0.1.3.2